<?xml version="1.0" encoding="UTF-8"?><!DOCTYPE article  PUBLIC "-//NLM//DTD Journal Publishing DTD v3.0 20080202//EN" "http://dtd.nlm.nih.gov/publishing/3.0/journalpublishing3.dtd"><article xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" dtd-version="3.0" xml:lang="en" article-type="research article"><front><journal-meta><journal-id journal-id-type="publisher-id">APM</journal-id><journal-title-group><journal-title>Advances in Pure Mathematics</journal-title></journal-title-group><issn pub-type="epub">2160-0368</issn><publisher><publisher-name>Scientific Research Publishing</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.4236/apm.2013.32037</article-id><article-id pub-id-type="publisher-id">APM-28587</article-id><article-categories><subj-group subj-group-type="heading"><subject>Articles</subject></subj-group><subj-group subj-group-type="Discipline-v2"><subject>Physics&amp;Mathematics</subject></subj-group></article-categories><title-group><article-title>
 
 
  &lt;i&gt;S&lt;/i&gt;&lt;sup&gt;1&lt;/sup&gt;-Equivariant CMC Surfaces in the Berger Sphere and the Corresponding Lagrangians
 
</article-title></title-group><contrib-group><contrib contrib-type="author" xlink:type="simple"><name name-style="western"><surname>eiichi</surname><given-names>Kikuchi</given-names></name><xref ref-type="aff" rid="aff1"><sub>1</sub></xref><xref ref-type="corresp" rid="cor1"><sup>*</sup></xref></contrib></contrib-group><aff id="aff1"><label>1</label><addr-line>Department of Mathematics, Tokai University, Tokyo, Japan</addr-line></aff><author-notes><corresp id="cor1">* E-mail:<email>kikuchi@jewel.ocn.ne.jp</email></corresp></author-notes><pub-date pub-type="epub"><day>05</day><month>03</month><year>2013</year></pub-date><volume>03</volume><issue>02</issue><fpage>259</fpage><lpage>263</lpage><history><date date-type="received"><day>October</day>	<month>4,</month>	<year>2012</year></date><date date-type="rev-recd"><day>November</day>	<month>30,</month>	<year>2012</year>	</date><date date-type="accepted"><day>December</day>	<month>17,</month>	<year>2012</year></date></history><permissions><copyright-statement>&#169; Copyright  2014 by authors and Scientific Research Publishing Inc. </copyright-statement><copyright-year>2014</copyright-year><license><license-p>This work is licensed under the Creative Commons Attribution International License (CC BY). http://creativecommons.org/licenses/by/4.0/</license-p></license></permissions><abstract><p>
 
 
   The periodic s<sup>1</sup>-equivariant hypersurfaces of constant mean curvature can be obtained by using the Lagrangians with suitable potential functions in the Berger spheres. In the corresponding Hamiltonian system, the conservation law is effectively applied to the construction of periodic s<sup>1</sup>-equivariant surfaces of arbitrary positive constant mean curvature. 
 
</p></abstract><kwd-group><kwd>&lt;i&gt;S&lt;/i&gt;&lt;sup&gt;1&lt;/sup&gt;-Equivariant CMC Surfaces; Conservation Laws</kwd></kwd-group></article-meta></front><body><sec id="s1"><title>1. Introduction</title><p>W.-Y. Hsiang [<xref ref-type="bibr" rid="scirp.28587-ref1">1</xref>] investigated the rotation hypersurfaces of constant mean curvature in the hyperbolic or spherical <img src="5-5300333\0e7c4df7-3bd9-470a-b826-8cf18f0c890a.jpg" />-space. In [<xref ref-type="bibr" rid="scirp.28587-ref2">2</xref>], Eells and Ratto have constructed the rotation (<img src="5-5300333\1f9c501f-3f7a-4504-8723-0a1ec5c9dde8.jpg" />-equivariant) minimal hypersurfaces in the unit 3-sphere with standard metric by using a certain first integral, which is invariant with respect to the rotation angle of generating curves on the orbit space. In [<xref ref-type="bibr" rid="scirp.28587-ref3">3</xref>], a family of <img src="5-5300333\0edffe03-0162-4d56-b41c-7e5f138878d8.jpg" />-equivariant periodic CMC surfaces was constructed in the Berger spheres when the constant mean curvature (CMC) is a sufficiently small positive number, and it was cleared that the conserved quantity can be obtained by using the Lagrangian equipped with suitable potential function of the corresponding dynamical system with respect to the Hsiang-Lawson metric [1,4] on the orbit space via the Hamilton equation, where the rotation angle of generating curves can be regarded as “time”. We should remark that the corresponding Lagrangian has the vanishing potential when we construct the <img src="5-5300333\46c749ef-30f5-4a3d-99eb-ee6c5c8dcde3.jpg" />-equivariant minimal hypersurfaces. However, in case that we construct the <img src="5-5300333\132f7179-aa33-4d2d-bc12-76a06e6491ba.jpg" />-equivariant non-minimal CMC-hypersurface in the Berger sphere, the potential of the Lagrangian is a nonvanishing function. In Theorem 4.3, we determine the potential function of the Lagrangian which corresponds to the <img src="5-5300333\068d28c4-f19b-43e5-8926-0a4d2cde044a.jpg" />-equivariant CMCsurfaces immersed in the Berger sphere. As a result we can obtain a family of periodic <img src="5-5300333\940abcff-e8ca-4cee-939f-c641770972d3.jpg" />-equivariant CMC surfaces in the Berger spheres when the constant mean curvature is an arbitrary positive number (Theorem 5.2).</p></sec><sec id="s2"><title>2. Preliminaries</title><p>In [<xref ref-type="bibr" rid="scirp.28587-ref3">3</xref>], a generalized inner product <img src="5-5300333\7d54e315-45a0-416f-bd59-6b322883431c.jpg" /> on the unit 3- sphere <img src="5-5300333\55781614-ed0b-420b-85fc-a04befb16d07.jpg" /> was defined by</p><p><img src="5-5300333\ee0451d3-830e-49dd-b179-c54ea49d2bb4.jpg" /></p><p>where<img src="5-5300333\75e2204e-1a40-4648-aa7f-05d1c430c2db.jpg" />,<img src="5-5300333\2ac88baa-49a0-4297-9ad2-64dedc73b7ac.jpg" /> and <img src="5-5300333\965b0165-9c78-45a6-8e99-52edcbf5929f.jpg" />, <img src="5-5300333\16b821a4-3eae-4336-80c7-5ad4d0d95fc3.jpg" />and <img src="5-5300333\aa9af983-c18a-4c96-8ce9-c06a9e684a2f.jpg" /> are positive and nonnegative parameters, respectively. The Cartan hypersurface <img src="5-5300333\ade9341c-6528-45cc-9bf8-fd243f1db7d2.jpg" /> in the unit 4-sphere is covered by <img src="5-5300333\ac70d697-f183-4d8a-921e-c577c8908051.jpg" /> (via an 8-fold covering), whose metric is rescaled along the Hopf fibres and its metric on <img src="5-5300333\0fda8c57-2d69-41a3-947f-03dae9c74aa2.jpg" /> coincides with <img src="5-5300333\1f36e8c5-a12e-43e3-be34-ed4830691414.jpg" /> [5,6]. The family of metrics <img src="5-5300333\3a2f95be-2dec-42a8-a3a9-c07abb769367.jpg" /> defined on <img src="5-5300333\24fcebf6-f187-4edc-bebd-98131c0ccff2.jpg" /> contains this one as a special case. In particular <img src="5-5300333\180b3555-f6ac-4c9c-b456-64e923f3789c.jpg" /> is a left-invariant metric on <img src="5-5300333\756c3728-b536-4b7e-90ea-d8a235828193.jpg" /> and <img src="5-5300333\71330db5-a38b-4fa8-9155-7bc7e2d0bebb.jpg" /> is called the Berger sphere with metric <img src="5-5300333\0ed389fe-7d54-482c-ad0a-f5acf98e2a48.jpg" />in case that<img src="5-5300333\936a335a-8179-4765-90cc-4cec4d4a5920.jpg" />. The Berger metrics <img src="5-5300333\b98f8300-fe8f-4eb4-be95-0d0b5e91e8b4.jpg" /> are obtained from the canonical metric by multiplying the metric along the Hopf fiber by <img src="5-5300333\0f0ec1d4-9026-4a96-9884-f74473fa72db.jpg" /> [<xref ref-type="bibr" rid="scirp.28587-ref7">7</xref>].</p><p>Throughout the paper we consider the Berger spheres<img src="5-5300333\ebc1343e-27a7-4507-b4a7-c92831ed06d0.jpg" />. Here we summarize the notations which are used in the paper.</p><p><img src="5-5300333\7d9b79d0-c05c-4780-a9ec-0f27f13d54ca.jpg" />denotes the orbit space by <img src="5-5300333\43c65a00-f006-4e24-aef8-4b6687fd33df.jpg" />-isometric <img src="5-5300333\1ff85c81-ba5d-487a-a9f4-abe03bfea9cb.jpg" />- ction <img src="5-5300333\0da5159d-fd16-4a55-aff1-389ee9052e07.jpg" /> as follows.</p><p><img src="5-5300333\3112d705-12c5-43d6-9d8d-d5c4eaddf2fd.jpg" /></p><p>As the parametrization of <img src="5-5300333\a37ac923-baff-4d5c-81ed-ffcccec0a54d.jpg" /> we use the following map:</p><p><img src="5-5300333\531de402-d988-4f29-9fba-bd0af558525f.jpg" /></p><p><img src="5-5300333\6317ecd6-f79c-4ec6-b407-931b6d042d19.jpg" />stands for the orbital metric on <img src="5-5300333\efec12f6-445c-4a08-b82a-53660f6f741c.jpg" />:</p><p><img src="5-5300333\01c2f974-6cb6-4138-a574-a12fc55dff2e.jpg" /></p><p><img src="5-5300333\b0516567-330f-4bf4-96f5-e43c4446c281.jpg" />is the volume function of orbits and <img src="5-5300333\62a45191-c36c-4d9c-a993-7261c4f66d4d.jpg" /> is the Hsiang-Lawson metric on<img src="5-5300333\398d30e1-be3a-4aaa-8a48-64be86a8c005.jpg" />:</p><p><img src="5-5300333\4828c0ef-05de-4e62-ad03-ea7b31422e3e.jpg" /></p><p>where</p><p><img src="5-5300333\3d80e917-4317-4b33-bcf3-d8a626406af0.jpg" /></p><p><img src="5-5300333\c26f52c8-ecf9-4685-afc5-1b5da5646e8d.jpg" />denotes a curve parametrized by arclength<img src="5-5300333\ca96632e-5858-421e-bc3c-8f417399b53e.jpg" />. And also <img src="5-5300333\bcff8fea-e0df-40ff-99d3-e8cc6b62013e.jpg" /> and <img src="5-5300333\c33687c6-878c-4aa2-9e66-1cb17ab89586.jpg" /> stand for the tension fields of <img src="5-5300333\11538506-105a-4c5f-9b5a-cad09f4c308f.jpg" /> with respect to the metrics <img src="5-5300333\cee42b92-8ac9-480f-b12b-7c19069cbf4a.jpg" /> and<img src="5-5300333\5c24c75a-2e2f-4892-ab5d-31ae0d6d1bb9.jpg" />, respectively. The geodesic curvature <img src="5-5300333\6c5947ef-13d4-4052-adc0-58542e13301a.jpg" /> at <img src="5-5300333\43cd7536-1bda-4645-bf53-3b85a88ac0d2.jpg" /> is defined by <img src="5-5300333\b825f6aa-9ecc-4be3-a597-55ef268c2349.jpg" /> where <img src="5-5300333\b1a7a0e2-cee6-4213-816f-3a8f3c336eb4.jpg" /> denotes the unit normal vector field to<img src="5-5300333\75137e05-6ce7-45a5-8c3a-5c2015a6a119.jpg" />.</p></sec><sec id="s3"><title>3. S<sup>1</sup>-Equivariant CMC-Immersion</title><p>For a curve<img src="5-5300333\5e161a4c-28e3-4049-915a-3561fe7f4ff5.jpg" />, we consider an <img src="5-5300333\351e4b25-f700-4ec6-9ea8-4381ae15518c.jpg" />-equivariant map <img src="5-5300333\4497dae7-6b84-4fc3-ba59-1443d4a30641.jpg" /> such that <img src="5-5300333\84f072f8-df32-46dc-8b9e-875f96d42551.jpg" />, where <img src="5-5300333\e7489c98-edf0-41ae-842c-89d2fec5fcf9.jpg" /> and <img src="5-5300333\4385a4c1-2a7b-4b85-a0be-35f77a2a74ff.jpg" /> are Riemannian submersions. Throughout the paper, we assume that <img src="5-5300333\f5626964-e830-463c-b088-7c58d76b4da5.jpg" /> is an <img src="5-5300333\133fc7ff-3e8f-4bd5-9411-8bb2d21377ce.jpg" />-equivariant constant mean curvature <img src="5-5300333\f49c827f-a088-4423-bcfb-39f07015a736.jpg" /> immersion. Then we have</p><disp-formula id="scirp.28587-formula110677"><label>(1)</label><graphic position="anchor" xlink:href="5-5300333\fffcecbf-5ffa-4113-817e-a16cc528dbd2.jpg"  xlink:type="simple"/></disp-formula><p>since</p><p><img src="5-5300333\881190a7-0bb5-4db2-a9b3-2542d636b835.jpg" /></p><p>On the orbit space<img src="5-5300333\b70f9b04-8391-4a96-9b42-74e951a2bf5b.jpg" />, the velocity vector field of a curve <img src="5-5300333\cb1d7e8a-1cff-4315-b500-4fc1412d2b8e.jpg" /> is given by the following component functions.</p><p><img src="5-5300333\765c46f0-bc5d-440b-a0dc-21a003b25b37.jpg" /></p><p>Lemma 3.1. The following formulas hold on <img src="5-5300333\33896267-119d-4d47-be83-2322ab094a35.jpg" />.</p><disp-formula id="scirp.28587-formula110678"><label>(2)</label><graphic position="anchor" xlink:href="5-5300333\bd0e3b59-56c9-419e-b1b8-82b37ef241ef.jpg"  xlink:type="simple"/></disp-formula><disp-formula id="scirp.28587-formula110679"><label>(3)</label><graphic position="anchor" xlink:href="5-5300333\26b21bc7-569c-4465-966c-1828220754d0.jpg"  xlink:type="simple"/></disp-formula><p>where</p><p><img src="5-5300333\da2b4d91-1327-49f2-861b-31c137147a41.jpg" /></p><p>and</p><p><img src="5-5300333\eae569d4-9fbd-4b2d-b207-6c468c381c1c.jpg" /></p><p>Then using the formula (1) we have the following differential Equation (4) of generating curves which corresponds to the CMC-rotation hypersurfaces immersed in<img src="5-5300333\33e952ca-be2f-4ccb-8640-4b499924a733.jpg" />, since using Lemma 3.1 the geodesic curvature <img src="5-5300333\bb86cd4c-a0e8-481b-b57e-334c5ef6de0e.jpg" /> is given by</p><disp-formula id="scirp.28587-formula110680"><label>(4)</label><graphic position="anchor" xlink:href="5-5300333\bc1a09a6-d80d-439d-879e-af8c5e787eeb.jpg"  xlink:type="simple"/></disp-formula></sec><sec id="s4"><title>4. Conservation Laws</title><p>We consider a generating curve <img src="5-5300333\83545929-9f39-4c42-a588-9b7593ad1870.jpg" /> on <img src="5-5300333\26e73090-2490-42f8-9ed4-a2c4a5c06e1b.jpg" /> such that <img src="5-5300333\0c888c28-95f9-4869-b6d9-984c2f83de20.jpg" /> and<img src="5-5300333\d436a87e-fcb0-41f9-a1d0-e73f133bb2d0.jpg" />. Then we can consider the space <img src="5-5300333\228f8209-dd48-4572-83fe-adf55f7853ab.jpg" /> of motion with <img src="5-5300333\ddbcafda-75d2-4bd9-bff4-8967c1f7d8c1.jpg" /></p><p>and time<img src="5-5300333\c40aae55-1163-4245-be56-55686dd7fca9.jpg" />. Let <img src="5-5300333\a0e0f9d4-7390-479b-9b89-14ad77ea632d.jpg" /> be a Lagrangian on<img src="5-5300333\4d2bb0b3-116d-4d3e-bd54-caabd6001438.jpg" />. Via the Legendre transformation we have the Hamiltonian <img src="5-5300333\8211373d-dfb2-4cf8-9022-a8b3927b3aa5.jpg" /> on the phase space<img src="5-5300333\b11ab2f7-16fe-4b93-b36c-fcbe2ddab1a4.jpg" />:</p><p><img src="5-5300333\0ddc73b8-ead4-4ccf-adec-6d58ac410746.jpg" /></p><p>The conservation laws of our system imply the following Proposition 4.1. Let the Lagrangian <img src="5-5300333\e3787b38-04fb-4771-9be5-beb3b12d551d.jpg" /> on <img src="5-5300333\043f6e62-031c-49ca-85d4-da0e4f45f282.jpg" /> be the following form:</p><p><img src="5-5300333\d82cc327-4386-4b3c-b0cf-aefcd758a43a.jpg" />where <img src="5-5300333\827e468a-307c-44bc-8d6f-824177a9840d.jpg" /> is the Hsiang-Lawson metric on <img src="5-5300333\84bab9a9-1d33-4d6d-a02b-ba3700dacfb3.jpg" /> and <img src="5-5300333\686123c6-632d-4af5-bc62-db8159da7853.jpg" /> is a potential function on the configuration space.</p><p>Then we have</p><disp-formula id="scirp.28587-formula110681"><label>(5)</label><graphic position="anchor" xlink:href="5-5300333\55472bfe-5b4c-4d9e-9881-88e19649b5c6.jpg"  xlink:type="simple"/></disp-formula><p>where the conserved quantity in the formula represents the Hamiltonian of our system.</p><p>By means of the Hamilton equation (5), we shall determine the potential <img src="5-5300333\c485b2e0-96c1-43ae-bbee-93312bcd1423.jpg" /> which corresponds to the <img src="5-5300333\ce87efa0-c6e5-48cc-a2e9-1ac9dfe40fbb.jpg" />-equivariant CMC surfaces immersed in <img src="5-5300333\3c6979bb-7718-4a8a-973f-b744ba0118fa.jpg" /> via the differential Equation (4) of generating curves on the orbit space<img src="5-5300333\8a8ddba0-7725-4f3b-8442-85b95d847341.jpg" />.</p><p>The direct computation yields the following Lemma 4.2. Assume that <img src="5-5300333\b3e4401e-259d-4509-a508-f21b71d37663.jpg" /> and <img src="5-5300333\58fbd068-7828-4e17-8b5c-99ca1bd7b031.jpg" /> are functions of</p><p><img src="5-5300333\e8c73f7c-bd1b-462b-84f6-d9b713eac03b.jpg" />and<img src="5-5300333\58bb0f6c-8e64-4f0d-b298-c8d5b8dfb058.jpg" />. Then we have</p><disp-formula id="scirp.28587-formula110682"><label>(6)</label><graphic position="anchor" xlink:href="5-5300333\2ee031df-6832-49bb-8f10-95ce943158c7.jpg"  xlink:type="simple"/></disp-formula><p>where</p><p><img src="5-5300333\6f9aa79f-f6bd-4e6e-959a-4cba18f69e98.jpg" /></p><p>As a consequence, we have the following Theorem 4.3. On our system, the Lagrangian <img src="5-5300333\de008667-82e7-42a7-b7db-0dc99c81bec7.jpg" /> and the Hamiltonian <img src="5-5300333\f064c60e-6555-4b26-af3a-48deb06428a3.jpg" /> which correspond to the <img src="5-5300333\7a710eb0-bec8-4fcc-8a14-0844904f6875.jpg" />-equivariant CMC-H hypersurface immersed in <img src="5-5300333\d605d92d-25a5-4663-bdf8-934a4b85cda8.jpg" /> can be determined as follows:</p><p><img src="5-5300333\72a42d77-7caa-4a5f-8f46-48ddb9e540c2.jpg" /></p><p>Proof. Using Lemma 4.2 and the differential equation of generating curves (4) we have</p><p><img src="5-5300333\7ef2c320-f46e-4b70-9239-918cf6271c04.jpg" /></p><p>from which we obtain</p><p><img src="5-5300333\ad6a2d08-5a55-404b-8f7b-a7e46065986e.jpg" /></p><p>Since <img src="5-5300333\8ceee326-d66f-41e0-8ba8-10012b7abbcb.jpg" /> is a constant mean curvature and</p><p><img src="5-5300333\0787f2f8-d36f-44a0-ba43-17a81001e402.jpg" /></p><p>we can choose such as<img src="5-5300333\39250271-6e10-4e56-affb-b15991bc1537.jpg" />. Q.E.D.</p></sec><sec id="s5"><title>5. Generating Curves for S<sup>1</sup>-Equivariant CMC Surfaces</title><p>Let <img src="5-5300333\01b052b5-dd19-4d03-ab11-447cdaa50d9f.jpg" /> be a generating curve on <img src="5-5300333\8f0c76e9-2540-4eec-ba8b-f3cf4bb9a839.jpg" /> such that <img src="5-5300333\5438449b-e4f5-4e63-89ec-7f07e962bfbc.jpg" /> and <img src="5-5300333\a1c34339-ca58-40bd-bb1d-335fc6b60aeb.jpg" /> with the arc length<img src="5-5300333\03baf3f2-c0e8-4e7c-943e-f4bb5c6fac9d.jpg" />. Then we set the following initial conditions:</p><p><img src="5-5300333\c74c7dd5-64c5-46a0-adeb-c5aee2fb68b4.jpg" /></p><p>The Hamilton equation <img src="5-5300333\60e3bafa-d334-41d5-86e3-54b4ab468698.jpg" /> (Theorem 4.3) implies that</p><p><img src="5-5300333\729d6a86-f3c0-4e8b-8de8-31f0afc1a4eb.jpg" /></p><p>from which we have</p><p><img src="5-5300333\22cb76a9-4c93-4bbb-b5b7-6cfac5269713.jpg" /></p><p>where</p><p><img src="5-5300333\0035866d-9a25-4186-9548-30d63578afee.jpg" /></p><p>On the other hand, using the formulas</p><p><img src="5-5300333\bf665b4f-dba3-4c4d-95a7-b0a26a4ed8ab.jpg" /></p><p>and</p><p><img src="5-5300333\10ee167e-141e-49b5-961c-dabdb221b338.jpg" /></p><p>we have</p><p><img src="5-5300333\bc8a682e-f963-4b1c-8a06-b2b8563c9a5e.jpg" /></p><p>Consequently we have the following Lemma 5.1. Under the initial conditions for generating curves which correspond to the CMC-H rotation hypersurfaces, we have</p><p><img src="5-5300333\162ad41e-aadf-4c9f-a1bd-f41d1c424441.jpg" /></p><p>and</p><p><img src="5-5300333\0c4a6975-481c-41e2-a411-aa5f662aa4ca.jpg" />(resp.,<img src="5-5300333\c59305db-6ef3-44c1-85a3-1888161de26d.jpg" />) if and only if,</p><p><img src="5-5300333\5c1b4826-cc15-479c-abd9-9380a05aea31.jpg" /></p><p>where</p><p><img src="5-5300333\bc64a802-cbff-4ff0-8c5b-c758b90325bd.jpg" /></p><p>Assume that <img src="5-5300333\f59060a6-ffa7-4eab-9a4d-7ecf05db145a.jpg" /> is an arbitrary positive number. In Lemma 5.1 we now choose <img src="5-5300333\350e7bda-4c60-4317-9be9-6ce818ce8276.jpg" /> such that</p><p><img src="5-5300333\a70fec03-6297-4d23-a8e9-05cf570a009f.jpg" />.</p><p>From Lemma 5.1, <img src="5-5300333\96f621fd-8d4c-4494-92da-8aead5c5451a.jpg" />and there exists the value <img src="5-5300333\8148710c-5545-4cd8-b6a9-a3bbab5ab3c6.jpg" /> of <img src="5-5300333\1e589e76-0f78-462f-8d18-8b2793d423da.jpg" /> such that <img src="5-5300333\e916a5db-f6a5-44f0-9a33-33fa9c36667a.jpg" /> decreases strictly until<img src="5-5300333\61b7809a-1cfb-41b6-a51e-95edf47573fb.jpg" />, where the value of <img src="5-5300333\6fd0991c-ed27-410e-8088-f255301d6b15.jpg" /> equals to zero at<img src="5-5300333\f4958ef8-cfd3-4f5e-bbd3-89333222ea75.jpg" />, and <img src="5-5300333\51c4ac4c-df4c-4fb3-90e4-29038396470b.jpg" /> takes a local minimum at<img src="5-5300333\c7e14251-7e9e-464d-87d3-1ca0a06fc5ed.jpg" />. In fact, if <img src="5-5300333\1796b039-57d8-497b-9198-b28c86b967ea.jpg" /> does not take a local minimum, then we may assume that there exists <img src="5-5300333\24413520-89e1-415a-9bc5-bd843a499e1e.jpg" /></p><p>such that <img src="5-5300333\38054801-3c22-4a35-b149-22c2b758bc5d.jpg" /> and</p><p><img src="5-5300333\4d701bab-56a1-4e9a-9d5c-0660f5c7ae69.jpg" />.</p><p>Then from the differential equation (4) of generating curves it follows that<img src="5-5300333\7e491351-5f54-4383-a8cc-8ba95918ce5f.jpg" />. On the other hand we obtain the following formula:</p><disp-formula id="scirp.28587-formula110683"><label>(7)</label><graphic position="anchor" xlink:href="5-5300333\025c8ca3-2590-401d-b792-2393f3137318.jpg"  xlink:type="simple"/></disp-formula><p>where</p><p><img src="5-5300333\a0e3813e-0c50-40f4-bda8-27331bce7a13.jpg" /></p><p>The formula (7) implies that</p><disp-formula id="scirp.28587-formula110684"><label>(8)</label><graphic position="anchor" xlink:href="5-5300333\7aedd7aa-36b7-46d7-87ad-b4c1914deb5c.jpg"  xlink:type="simple"/></disp-formula><p>where</p><p><img src="5-5300333\a3becfe0-b457-4377-8e13-3687f9c5958f.jpg" /></p><p>The formula <img src="5-5300333\de506063-094e-40a6-a825-ff7d52ce4282.jpg" /> implies that</p><p><img src="5-5300333\8eefeedd-d44f-4c2f-a3a4-8ae6273fdf0d.jpg" /></p><p>from which we have</p><p><img src="5-5300333\c6bcf2c8-0718-45ce-bf46-b7cc477f1621.jpg" /></p><p>since<img src="5-5300333\9a6540f6-0fba-4e56-ad29-5942a6d748dd.jpg" />.</p><p>Hence we see that<img src="5-5300333\6ee711aa-397a-4af8-9dd5-2d35db779c47.jpg" /> is a positive number. Now if<img src="5-5300333\73df3b8c-0eda-48bc-96fd-7b7cbe9473d6.jpg" />, then<img src="5-5300333\3e4f67fc-3c67-42ac-afa0-678a2130d0c8.jpg" />, which implies that <img src="5-5300333\7652f484-b31b-4ad1-bfac-ea6c35d559e7.jpg" /> and</p><p><img src="5-5300333\9ca080ee-e2d7-4e0a-87b6-8da46b4cec50.jpg" />, hence<img src="5-5300333\0b2647f5-6c73-4899-a00d-199e7619da05.jpg" />, which is a contradiction. Therefore, the value <img src="5-5300333\8954ebd1-177e-44fc-8ef5-093f23b10ea1.jpg" /> is not zero.</p><p>Consequently, since<img src="5-5300333\f1e1961d-9248-41ec-bb02-5fa4cd98bc2a.jpg" />, from the formula (8)</p><p>we see that <img src="5-5300333\2de09325-e6c5-4ebd-9018-306a7e2fb606.jpg" />is not zero, which contradicts the assumption<img src="5-5300333\a58497b6-21b4-4bdb-8eb8-6b2e4cd7cb1a.jpg" />. Hence <img src="5-5300333\2a5989e6-e82b-40be-84ab-c02abcf453bf.jpg" /></p><p>takes a local minimum.</p><p>Thus we can continue<img src="5-5300333\c03862ac-94bb-40fd-a439-ba7d931de106.jpg" /> as the curve satisfying the differential Equation (4) by the reflection. Let <img src="5-5300333\f2b1f674-9f26-44c9-ba5b-988f308f11df.jpg" /> be the right hand side of (7). We can define <img src="5-5300333\3ed0d6da-5675-4215-88f6-78a6291fa0e1.jpg" /> by <img src="5-5300333\28b80183-6e1e-41c8-b01a-7b4695fded76.jpg" /> as follows:</p><p><img src="5-5300333\b4d6c3c2-96e5-4689-bb5d-70a794e70fdf.jpg" /></p><p>Consequently we have the following Theorem 5.2. Let <img src="5-5300333\0eacb503-14cc-4c0d-8ee4-36b785b943ef.jpg" /> be an arbitrary positive number and choose <img src="5-5300333\9dbe0084-c747-466e-bd32-05c64d4fc837.jpg" /> such that<img src="5-5300333\1f256c12-bd7b-4670-9824-1df6a59e6c2d.jpg" />. If <img src="5-5300333\bbea1b9f-b2e0-42fc-abf7-307838c67cbd.jpg" /> is a rational number, then the corresponding <img src="5-5300333\7aaef5f8-5459-48bc-bc38-5561ca2dec9b.jpg" />-equivariant hypersurface is an immersed CMC-H torus in the Berger sphere<img src="5-5300333\2146e414-d22c-4c10-8d9e-565621892ca0.jpg" />. In particular, if <img src="5-5300333\7be6eccc-5b94-4aab-9281-a01f2fe864cc.jpg" />is an integer, then this CMC-H torus is embedded.</p><p>Theorem 5.3. In the case<img src="5-5300333\f2155b92-cf90-417a-93a6-f45a788ddc17.jpg" />, Then the corresponding <img src="5-5300333\717c1cb0-9096-4320-90f3-497d6373f778.jpg" />-equivariant CMC-H hypersurface in the Berger sphere <img src="5-5300333\6f03a00b-773a-46ca-ba4a-f7207f81222d.jpg" /> is an extended Clifford torus</p><p><img src="5-5300333\67e309fb-254a-496e-a1b0-a1458ae557c7.jpg" /></p><p>where</p><p><img src="5-5300333\eb79e114-34d3-4c33-aa72-f0ff632f188b.jpg" /></p><p>Corollary 5.4. There exists an embedded minimal torus in the Berger sphere <img src="5-5300333\1baa8478-c820-4d24-a8fd-d26d4815ff61.jpg" /></p><p><img src="5-5300333\639875f2-56d5-4c10-888f-f2bc3ef10fc4.jpg" /></p></sec><sec id="s6"><title>6. Acknowledgements</title><p>I am grateful to Yoshihiro Ohnita and Junichi Inoguchi for their encouragement.</p></sec><sec id="s7"><title>REFERENCES</title></sec></body><back><ref-list><title>References</title><ref id="scirp.28587-ref1"><label>1</label><mixed-citation publication-type="other" xlink:type="simple">W-Y. Hsiang, “On Generalization of Theorems of A. D. Alexandrov and C. Delaunay on Hypersurfaces of Constant Mean Curvature,” Duke Mathematical Journal, Vol. 49, No. 3, 1982, pp. 485-496.  
doi:10.1215/S0012-7094-82-04927-4</mixed-citation></ref><ref id="scirp.28587-ref2"><label>2</label><mixed-citation publication-type="other" xlink:type="simple">J. Eells and A. Ratto, “Harmonic Maps and Minimal Immersions with Symmetries,” Annals of Mathematics Stu- dies, No. 130, 1993.</mixed-citation></ref><ref id="scirp.28587-ref3"><label>3</label><mixed-citation publication-type="other" xlink:type="simple">K. Kikuchi, “The Construction of Rotation Surfaces of Constant Mean Curvature and the Corresponding Lagrangians,” Tsukuba Journal of Mathematics, Vol. 36, No. 1, 2012, pp. 43-52.</mixed-citation></ref><ref id="scirp.28587-ref4"><label>4</label><mixed-citation publication-type="other" xlink:type="simple">W-Y. Hsiang and H. B. Lawson, “Minimal Submanifolds of Low Cohomogeneity,” Journal of Differential Geometry, Vol. 5, 1971, pp. 1-38.</mixed-citation></ref><ref id="scirp.28587-ref5"><label>5</label><mixed-citation publication-type="other" xlink:type="simple">D. Ferus and U. Pinkall, “Constant Curvature 2-Spheres in the 4-Sphere,” Mathematische Zeitschrift, Vol. 200, No. 2, 1989, pp. 265-271. doi:10.1007/BF01230286</mixed-citation></ref><ref id="scirp.28587-ref6"><label>6</label><mixed-citation publication-type="other" xlink:type="simple">H. Muto, Y. Ohnita and H. Urakawa, “Homogeneous Minimal Hypersurfaces in the Unit Spheres and the First Eigenvalues of Their Laplacian,” Tohoku Mathematical Journal, Vol. 36, No. 2, 1984, pp. 253-267.  
doi:10.2748/tmj/1178228851</mixed-citation></ref><ref id="scirp.28587-ref7"><label>7</label><mixed-citation publication-type="other" xlink:type="simple">P. Petersen, “Riemannian Geometry,” Graduate Texts in Mathematics, 2nd Edition, Vol. 171, Springer-Verlag, New York, 2006.</mixed-citation></ref></ref-list></back></article>